One of the most heavily investigated analyte sensing devices is an implantable glucose sensor for detecting glucose levels in patients with diabetes. Despite the increasing number of individuals diagnosed with diabetes and recent advances in the field of implantable glucose monitoring devices, currently used devices are unable to provide data safely and reliably for long periods of time (e.g., months or years) [See, e.g., Moatti-Sirat et al., Diabetologia 35:224-30 (1992)]. There are two commonly used types of implantable glucose sensing devices. These types are those which are implanted intravascularly and those implanted in tissue.
With reference to devices that may be implanted in tissue, a disadvantage of these devices has been that they tend to lose their function after the first few days to weeks following implantation. At least one reason for this loss of function has been attributed to the fact that there is no direct contact with circulating blood to deliver sample to the tip of the probe of the implanted device. Because of these limitations, it has previously been difficult to obtain continuous and accurate glucose levels. However, this information is often extremely important to diabetic patients in ascertaining whether immediate corrective action is needed in order to adequately manage their disease.
Some medical devices, including implanted analyte sensors, drug delivery devices and cell transplantation devices require transport of solutes across the device-tissue interface for proper function. These devices generally include a membrane, herein referred to as a cell-impermeable membrane that encases the device or a portion of the device to prevent access by host inflammatory or immune cells to sensitive regions of the device.
A disadvantage of cell-impermeable membranes is that they often stimulate a local inflammatory response, called the foreign body response (FBR) that has long been recognized as limiting the function of implanted devices that require solute transport. Previous efforts to overcome this problem have been aimed at increasing local vascularization at the device-tissue interface with limited success.
The FBR has been well described in the literature and is composed of three main layers, as illustrated in FIG. 1. The innermost FBR layer 40, adjacent to the device, is composed generally of macrophages and foreign body giant cells 41 (herein referred to as the barrier cell layer). These cells form a monolayer 40 of closely opposed cells over the entire surface 48a of a smooth or microporous (<1.0 .mu.m) membrane 48. The intermediate FBR layer 42 (herein referred to as the fibrous zone), lying distal to the first layer with respect to the device, is a wide zone (30-100 microns) composed primarily of fibroblasts 43 and fibrous matrix 44. The outermost FBR layer 46 is loose connective granular tissue containing new blood vessels 45 (herein referred to as the vascular zone 46). A consistent feature of the innermost layers 40 and 42 is that they are devoid of blood vessels. This has led to widely supported speculation that poor transport of molecules across the device-tissue interface 47 is due to a lack of vascularization near interface 47 (Scharp et al., World J. Surg. 8:221-229 (1984), Colton and Avgoustiniatos J. Biomech. Eng. 113:152-170 (1991)).
Patents by Brauker et al. (U.S. Pat. No. 5,741,330), and Butler et al. (U.S. Pat. No. 5,913,998), describe inventions aimed at increasing the number of blood vessels adjacent to the implant membrane (Brauker et al.), and growing within (Butler et al.) the implant membrane at the device-tissue interface. The patent of Shults et al. (U.S. Pat. No. 6,001,067) describes membranes that induce angiogenesis at the device-tissue interface of implanted glucose sensors. FIG. 2 illustrates a situation in which some blood vessels 45 are brought close to an implant membrane 48, but the primary layer 40 of cells adherent to the cell-impermeable membrane blocks glucose. This phenomenon is described in further detail below.
In the examples of Brauker et al. (supra), and Shults et al., bilayer membranes are described that have cell impermeable layers that are porous and adhesive to cells. Cells are able to enter into the interstices of these membranes, and form monolayers on the innermost layer, which is aimed at preventing cell access to the interior of the implanted device (cell impenetrable layers). Because the cell impenetrable layers are porous, cells are able to reach pseudopodia into the interstices of the membrane to adhere to and flatten on the membrane, as shown in FIGS. 1 and 2, thereby blocking transport of molecules across the membrane-tissue interface. The known art purports to increase the local vascularization in order to increase solute availability. However, the present studies show that once the monolayer of cells (barrier cell layer) is established adjacent to the membrane, increasing angiogenesis is not sufficient to increase transport of molecules such as glucose and oxygen across the device-tissue interface.
One mechanism of inhibition of transport of solutes across the device-tissue interface that has not been previously discussed in the literature is the formation of a uniform barrier to analyte transport by cells that form the innermost layer of the foreign body capsule. This layer of cells forms a monolayer with closely opposed cells having tight cell-to-cell junctions. When this barrier cell layer forms, it is not substantially overcome by increased local vascularization. Regardless of the level of local vascularization, the barrier cell layer prevents the passage of molecules that cannot diffuse through the layer. Again, this is illustrated in FIG. 2 where blood vessels 45 lie adjacent to the membrane but glucose transport is significantly reduced due to the impermeable nature of the barrier cell layer 40. For example, both glucose and its phosphorylated form do not readily transit the cell membrane and consequently little glucose reaches the implant membrane through the barrier layer cells.
It has been confirmed by the present inventors through histological examination of explanted sensors that the most likely mechanism for inhibition of molecular transport across the device-tissue interface is the barrier cell layer adjacent to the membrane. There is a strong correlation between desired device function and the lack of formation of a barrier cell layer at the device-tissue interface. In the present studies, devices that were observed histologically to have substantial barrier cell layers were functional only 41% of the time after 12 weeks in vivo. In contrast, devices that did not have significant barrier cell layers were functional 86% of the time after 12 weeks in vivo.
Consequently, there is a need for a membrane that interferes with the formation of a barrier layer and protects the sensitive regions of the device from host inflammatory response.